ENCRYPTED TRAFFIC CLASSIFICATION BASED ON NETWORK FLOW TIME-SPACE SERIES
Traffic classification is very important for network resource management and security.However,user traffic is often encrypted,which brings great challenges to traffic classification.Therefore,we propose a novel time series feature extraction technique to address the encrypted traffic classification problem.We extracted significant attributes of the encrypted network traffic behavior by analyzing the time series of received packets.We used the LSTM combined with attention mechanism to train and classify traffic.To evaluate the efficiency of the proposed method,we carried out intensive experiments on an open network dataset ISCXVPN2016.The experimental results show that the proposed method can significantly improve the performance in identifying encrypted application traffic in terms of accuracy and computation efficiency.
Deep learningEncrypted traffic classificationNeutral network
万方数据
维普
